Answers are grounded in this paper and cite their sources.
Cox and Snell have pointed out that when the empirical variance of a measured response y is quite different from its nominal variance under an assumed probability model, standard errors that are not adjusted for the discrepancy may yield misleading inferences. This article describes a test for assessing departures from nominal dispersion in survival analysis and case-control studies. Also, a robust procedure for adjusting standard errors for these models is given.
No takes yet. Share an insight, caveat, or question.
Philip J. Smith (1993) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: